The longform trap

By now, many of you have probably heard about the uproar surrounding an article that was published two weeks ago by the sports site SB Nation. For those who missed it, here’s a quick summary: Jeff Arnold, a freelance journalist whose work has appeared in the New York Times, wrote a 12,000-word profile of the former Oklahoma City policeman Daniel Holtzclaw, who was convicted in December of multiple counts of rape and sexual battery of women he had detained while on duty. The piece, titled “Who is Daniel Holtzclaw?,” focused on the subject’s collegiate football career, which Arnold had covered in Michigan. Within hours of the story’s appearance, a growing furor mounted online over its many inexplicable aspects: it took a weirdly sympathetic tone toward a man responsible beyond a reasonable doubt for reprehensible crimes, relied almost entirely on sources inclined in Holtzclaw’s favor, and repeatedly cast aspersions on the characters of his victims. In its readiness to emphasize the defense case and the words of those who believed in Holtzclaw’s innocence, it seemed poised to attempt a Making a Murderer, only to fall lamely back onto the qualification that yes, he was guilty, but let’s talk about something else instead. And its muddled final paragraph gives a good sense of its confusions:

Pending an improbable successful appeal, everything [Holtzclaw] had worked for was now gone, likely never to be recovered, ever again. Recovery, if there is any, appears to be something deserved only by the victims of a man whose belief in his innocence will apparently be, like the way he once pursued his dream of playing in the NFL, unrelenting, despite all evidence to the contrary.

The backlash was swift and furious, and SB Nation quickly pulled the entire piece. Editorial director Spencer Hall—who had read the article before its release—posted a note calling its publication “a complete breakdown of a part of the editorial process at SB Nation” and characterizing the story itself as “a complete failure.” Its editor, Glenn Stout, was fired, and the site broke all ties with Arnold, although the damage was already done. And the only silver lining to the whole fiasco was the coverage of the incident by Deadspin, which jumped on it right away and followed up with an excellent analysis of what went wrong. That postmortem, written by Greg Howard, ought to be read by everybody who cares about the challenges confronting online journalism today, as brought into stark relief by a moment in which the system collapsed entirely. And one statement in particular deserves to be memorized by all writers, editors, and readers:

There is no such thing as longform writing. There is such a thing as features writing—profiles, investigations, essays—and if it’s prestigious, that’s mainly because of its association with careful selection of subjects and with vigorous research, reporting, editing, copy-editing, and fact-checking. A feature carries an implicit assertion that a publication has invested money, time, talent, effort, and care to produce something of depth. Longform is a variant of feature writing—a branding strategy, really—that confuses a secondary indicator (length) for the thing itself (quality). As the name implies, it asserts nothing more than that a certain mass has been attained.

I’ve been mulling over this observation ever since. The profile made a lot of mistakes, and Howard shrewdly notes that it committed the basic error of treating the people around Holtzclaw as sources instead of subjects. But above all else, it was the result of an inability to distinguish between length and quality. The story was produced by a “vertical” designed explicitly to produce long articles, which are perceived as lending prestige to sites that otherwise focus on short pieces: the implication is that time and resources are being devoted to the development of meaningful journalism. Unfortunately, length might be a convenient heuristic, but it isn’t a guarantee, especially when a site mandates a certain word count before a story gets the prestige treatment, which is exactly the opposite of how it should work. (Howard writes: “One freelancer said that per the terms of the contract, the story [for SB Nation Longform] had to be at least 4,000 words long.”) And if that confusion persists, it’s because it’s easier than evaluating a story on its real merits. A few years back, the New York Times article “Snow Fall” set a new standard for the presentation of long works of reportage online, and countless pieces ever since have copied its look, but not its level of craftsmanship. With its splashy header image, pullout quotes, tastefully integrated photos, and endless wall of text, “Who is Daniel Holtzclaw?” looked exactly the way such stories are supposed to look—all it was missing was parallax scrolling. But it wasn’t a good piece. It was a simulation of one.

And it’s tempting, even for other writers, to judge and share stories based solely on that surface sheen. As Barry Petchesky of Deadspin noted in an earlier piece on the same subject: “These are the pieces that people praise without reading.” We’ve all been guilty of this: I’ve often linked to articles on this blog based on little more than a cursory skim to verify that they look like reputable sources, which means that I’m part of the problem. Last year, I wrote a pair of posts titled “The AutoContent Wizard,” in which I noted that the majority of listicles or slideshows are a kind of content mirage, cranked out for the convenience of the creator rather than the enjoyment of the audience. When I look back, though, I think I missed a key point by failing to draw a similar conclusion about longreads: the pursuit of length for its own sake can also take the place of reasoned critical judgment. (It reminds me of my argument that Apple has turned thinness and lightness into proxies for innovation, since they’re easier to measure and quantify, even if they lack a meaningful relationship to the user experience.) Overworked editors, under enormous pressure to produce content on a regular basis, aren’t going to be inclined to kill a piece that has already consumed time and effort. And such a situation will lead to occasional implosions. As Petchesky concluded:

There had never been a complete failure of concept and execution quite like this one, but it was nearly inevitable. If a company has a gorgeous CMS designed for longform, and a mandate to produce longform, and staff in place to present longform, it’s going to publish longform—whether the stories are there or not.

The only solution is for readers and editors to conscientiously distinguish between legitimate journalism and the mimics, like cryptic insects, that pose as the real thing. It won’t be easy. But it sure beats the alternative.